DESCRIPTION: (Principal Investigator's Abstract) The long term goal of this project continues to be the production of analogs of rifamycin and evaluation of their activity as antitubercular drugs. We hope to develop forms with activity against Mycobacterium sp. that have become resistant to rifampicin and(or) with improved potency over this drug and the three other rifamycin analogs in current use clinically. Our working hypothesis is based primarily on elimination of functionalities that are modified by the three types of resistance mechanisms used by M. tuberculosis, but we also envisage the possibility of altering the drug's binding to the RNA polymerase beta-subunit, which is the most prevalent resistance mechanism in other bacteria. The analogs will be prepared by combinatorial biosynthesis, using modified polyketide synthase genes. The following specific aims will be pursued in the Hutchinson and Khosla laboratories at Wisconsin and Stanford, respectively, in collaboration with the Floss laboratory at Washington. (1) Develop a collection of Amycolatopsis mediterranei strains with deletions in different PKS genes suitable for expression of individual PKS genes and reconstitution of active PKS in vivo. (2) Establish the minimum number of genes required to form proansamycin X or rifamycin W in A. mediterranei and in a heterologous host such as Streptomyces lividans. (3) Explore the behavior of the Rif PKS in vivo and in vitro, in an attempt to understand why the intermediates of polyketide chain assembly are spontaneously released from the enzyme subunits at a high frequency. (4) Systematically modify the rif PKS genes by domain and module replacement targeted at altering position 7 of the naphthoquinone ring and positions 21 to 25 in the ansamycin ring. (5) Engineer the rifamycin gene cluster for production of polyketides related to geldanamycin through the use of hybrid PKSs made from rif and ery genes. (6) Engineer the rifamycin gene cluster for overproduction of rifamycin B and eventual rifamycin analogs. (7) If necessary, move some of the rifamycin gene cluster into Streptomyces lividans to facilitate aims 4 - 6.